PROJECT SUMMARY This proposal requests five-years of support for studies of binaural (two-ear) processing, with particular emphasis on listening to speech in multisource and reverberant environments. The primary motivation for this work is the difficulty some people experience when listening to speech in noisy environments. The work is primarily concerned with cases when multiple talkers are speaking at the same time as the target speaker and/or cases that have significant reflections off surfaces. It is often observed that abilities in these environments are compromised when one or both ears are not fully functioning and with the advancement of age; thus, the focus on the binaural hearing contributions to our abilities in difficult environments and the effects of aging. The proposed work includes a blend of experimental measurements and computational modeling of the binaural and spatial advantages in this cocktail party environment. The measurement program is designed to separate the multiple factors that affect performance in this complex task, and the modeling both determines the degree to which current understanding of two-eared processing can explain empirical abilities and attempts to extend our understanding. Thus, we will be exploiting our history of binaural modeling at the signal-processing or black-box level as well as at the neurophysiological level. Work in all of these areas will include both normal hearing and impaired hearing measurements and models and will include the measurement and analysis of performance with hearing aids. This study includes specific attention to the distinctive difficulties that are often observed in elderly listeners. This work should lead to improved hearing aids as well as better understanding of the process of hearing in complex environments.